The economics of the village gossip

Psst. Want to know who the best gossip in the village is?

It turns out there’s a formula for that. Better yet: word on the street is you can avoid the complicated economics and just ask the villagers instead.

Economists Abhijit Banerjee and Esther Duflo have in recent years been increasingly acclaimed for the rigorous randomised trials on esoteric subjects that they and their researchers conduct at the Massachusetts Institute of Technology’s Abdul Latif Jameel Poverty Action Lab.

Their provocative 2011 book “Poor Economics” served as big poke in the eye of the development community and the way it worked. Aid workers, they argued, too often ignore the real – and flawed – decisions that poor people make and, more importantly, the odd incentives they are presented with. Why, for example, do impoverished Moroccan farmers consider buying a TV more important than food? Or Indonesian villagers teetering on the financial edge choose to have so many children? Or mothers in Africa risk their children contracting malaria (and having to pay an expensive medical bill) rather than spend a small amount on a mosquito net?

But in their latest working paper, published last month with Stanford University colleagues Arun Chandrasekhar and Matthew Jackson, they have tackled a subject with arguably more universal appeal: gossip. And the very important question of just how to identify the biggest gossip in the village.

First things first: gossip matters.

For policy makers and businesses, getting the right bit of information out and doing so efficiently can make a huge difference. Whether you are spreading information about an immunisation campaign or a new product you want to make sure you do so in the most efficient way possible.

In a study of 43 villages in India’s Karnataka province published last year in the journal Science, Banerjee, Duflo, and their Stanford colleagues looked at how information about microfinance travelled. They discovered that the best conduits for information were not necessarily the traditional leaders in villages but rather the gossips.

They even came up with a formula to measure the “diffusion centrality” of individuals. Call it the gossip index. It looks like this (see the paper for an explanation of its constituent parts!):

But Banerjee says they also realised very quickly that identifying the best gossip in town the way the economists set out to is a completely impractical exercise. In the case of the microfinance study, it took them years of data collection and analysis and it is extremely unlikely any business would choose to do the same.

“If you had to figure out the entire social network, it would be very expensive to do it,” Banerjee told me during a phone interview from New Delhi this week.

So, cue the latest study. Conducted in 35 villages in India, it tested whether there might be an easier way to identify the best gossips. The economists used the data on the existing social networks to identify those with high gossip index scores. Then they asked villagers two questions:

(Loan) If we want to spread information about a new loan product to everyone in your village, to whom do you suggest we speak?

(Event) If we want to spread information to everyone in the village about tickets to a music event, drama, or fair that we would like to organize in your village, to whom should we speak?

Half the villagers they spoke to were much too polite and so declined to answer those questions. They didn’t want to gossip about the gossips.

But those who did were remarkably in agreement about who they picked. Their answers were focused on a particular 5 per cent of the community in the case of the loan question and 4 per cent in the case of the event.

Moreover, the individuals they picked were people who generally scored highly on the gossip index. And, as was the case in last year’s microfinance study, the best conduits for information were not necessarily the formal leaders in villages.

The findings shouldn’t be that surprising. We all know who the office/village gossips are and so the results are an affirmation of the power of human intuition.

“We know who the gossip here is. It’s no more than that,” Banerjee told me. “But [the study] does say that your sense of who is the office gossip is generally very good and that if you want to spread information you should do it by that person.”

There is, as we flagged above, a lesson for businesses and policy makers in that. When you want to get the word out about something new it is always best to try to do so through the village gossip. And, it turns out that we all generally know who that is. Just ask.

Back to beyondbrics